Scientific AI is rapidly transforming fundamental physics research and challenging defining aspects of the fundamental physics methodology. We discuss opportunities and dangers of this transformation and find exciting benefits from a close interaction between AI and fundamental physics, provided that we remain aware of the scientific methodologies of the respective fields. For fundamental physics, we discuss two such aspects: statistical validation and a generalizing theory description, both with the goal of discovering new physics in vast datasets.

Germany

Michael Krämer

Tilman Plehn

Machine learning is good for physics—and vice versa

Over the last decade, non-relativistic theories have attracted considerable attention. In general, such theories can be obtained by contracting relativistic parent theories. In this work, we discuss features of non-relativistic supersymmetric field theories from both the Galilean and Carrollian points of view that may be useful for constructing electric and magnetic non-relativistic theories.

Türkiye

Osman Ergec

Aspects of non-relativistic supersymmetric theories

We investigate <italic>X</italic>(6900) as a hadronic realization of a gluonic portal between a dark sector and QCD. The analysis is updated using the 2026 CMS observation of a family of all-charm structures and a recent combined LHCb&#8211;ATLAS&#8211;CMS analysis of the di- <italic>J</italic>/ <italic>&#968;</italic> spectrum. The CMS angular analysis favors <math altimg='si19.svg'><mrow><msup><mi>J</mi><mrow><mi>P</mi><mi>C</mi></mrow></msup><mo linebreak='goodbreak'>=</mo><msup><mn>2</mn><mrow><mo>+</mo><mo>+</mo></mrow></msup></mrow></math>, motivating a spin-two mediator rather than the scalar ansatz used in earlier versions of this study. The combined analysis also demonstrates that the extracted <italic>X</italic>(6900) mass and width are significantly model dependent because of interference. We therefore use <math altimg='si20.svg'><mrow><msub><mi>M</mi><mi>X</mi></msub><mo linebreak='goodbreak'>=</mo><mn>6.919</mn><mspace width='0.33em' /><mtext>GeV</mtext></mrow></math> and <math altimg='si21.svg'><mrow><msub><mstyle mathvariant='normal'><mi>&#915;</mi></mstyle><mi>X</mi></msub><mo linebreak='goodbreak'>=</mo><mn>70.3</mn><mspace width='0.33em' /><mtext>MeV</mtext></mrow></math> as a reference benchmark, corresponding to the narrow-width, weak-interference Model I of the combined analysis, while treating the spread among interference models as a systematic theoretical uncertainty. The interaction is <display><formula id='ueq0001'><math altimg='si22.svg'><mrow><msub><mi mathvariant='script'>L</mi><mrow><mrow><mi mathvariant='normal'>i</mi></mrow><mi>n</mi><mi>t</mi></mrow></msub><mo linebreak='goodbreak'>=</mo><mo>&#8722;</mo><mfrac><msub><mi>c</mi><mi>&#967;</mi></msub><mstyle mathvariant='normal'><mi>&#923;</mi></mstyle></mfrac><msub><mi>X</mi><mrow><mi>&#956;</mi><mi>&#957;</mi></mrow></msub><msubsup><mi>T</mi><mi>&#967;</mi><mrow><mi>&#956;</mi><mi>&#957;</mi></mrow></msubsup><mo linebreak='goodbreak'>&#8722;</mo><mfrac><msub><mi>c</mi><mi>g</mi></msub><mstyle mathvariant='normal'><mi>&#923;</mi></mstyle></mfrac><msub><mi>X</mi><mrow><mi>&#956;</mi><mi>&#957;</mi></mrow></msub><msubsup><mi>T</mi><mi>g</mi><mrow><mi>&#956;</mi><mi>&#957;</mi></mrow></msubsup><mo>.</mo></mrow></math></formula></display>Using the standard spin-two normalization, <display><formula id='ueq0002'><math altimg='si23.svg'><mrow><mstyle mathvariant='normal'><mi>&#915;</mi></mstyle><mrow><mo>(</mo><mi>X</mi><mo>&#8594;</mo><mi>g</mi><mi>g</mi><mo>)</mo></mrow><mo linebreak='goodbreak'>=</mo><mfrac><mrow><msubsup><mi>c</mi><mi>g</mi><mn>2</mn></msubsup><msubsup><mi>M</mi><mi>X</mi><mn>3</mn></msubsup></mrow><mrow><mn>10</mn><mi>&#960;</mi><msup><mstyle mathvariant='normal'><mi>&#923;</mi></mstyle><mn>2</mn></msup></mrow></mfrac><mo>,</mo></mrow></math></formula></display>and the gluonic gravitational form factor <italic>A<inf>g</inf></italic>(0), the leading spin-independent nucleon cross section is <display><formula id='ueq0003'><math altimg='si24.svg'><mrow><msubsup><mi>&#963;</mi><mrow><mi>&#967;</mi><mi>N</mi></mrow><mrow><mrow><mi mathvariant='normal'>S</mi></mrow><mi>I</mi></mrow></msubsup><mo linebreak='goodbreak'>=</mo><mfrac><mrow><msubsup><mi>&#956;</mi><mrow><mi>&#967;</mi><mi>N</mi></mrow><mn>2</mn></msubsup><msubsup><mi>m</mi><mi>&#967;</mi><mn>2</mn></msubsup><msubsup><mi>m</mi><mi>N</mi><mn>2</mn></msubsup></mrow><mi>&#960;</mi></mfrac><msup><mrow><mo stretchy='true'>(</mo><mfrac><mrow><msub><mi>c</mi><mi>&#967;</mi></msub><msub><mi>c</mi><mi>g</mi></msub></mrow><mrow><msup><mstyle mathvariant='normal'><mi>&#923;</mi></mstyle><mn>2</mn></msup><msubsup><mi>M</mi><mi>X</mi><mn>2</mn></msubsup></mrow></mfrac><mo stretchy='true'>)</mo></mrow><mn>2</mn></msup><msub><mi>A</mi><mi>g</mi></msub><msup><mrow><mo>(</mo><mn>0</mn><mo>)</mo></mrow><mn>2</mn></msup><mo>.</mo></mrow></math></formula></display>For <math altimg='si2.svg'><mrow><msub><mi>A</mi><mi>g</mi></msub><mrow><mo>(</mo><mn>0</mn><mo>)</mo></mrow><mo linebreak='goodbreak'>=</mo><mn>0.42</mn></mrow></math>, <math altimg='si8.svg'><mrow><msub><mi>B</mi><mrow><mi>g</mi><mi>g</mi></mrow></msub><mo linebreak='goodbreak'>=</mo><mn>0.10</mn></mrow></math>, and <math altimg='si25.svg'><mrow><msub><mi>m</mi><mi>&#967;</mi></msub><mo linebreak='goodbreak'>=</mo><mn>100</mn><mspace width='0.33em' /><mtext>GeV</mtext></mrow></math>, the benchmark reference-scale estimate implies <math altimg='si26.svg'><mrow><msub><mi>c</mi><mi>&#967;</mi></msub><mo linebreak='goodbreak'>/</mo><mstyle mathvariant='normal'><mi>&#923;</mi></mstyle><mo>&#8818;</mo><mn>1.87</mn><mo linebreak='goodbreak'>&#215;</mo><msup><mn>10</mn><mrow><mo>&#8722;</mo><mn>8</mn></mrow></msup><mspace width='0.33em' /><msup><mrow><mtext>GeV</mtext></mrow><mrow><mo>&#8722;</mo><mn>1</mn></mrow></msup></mrow></math> when the 90% C.L. XENONnT reference scale is imposed. The annihilation channel <math altimg='si4.svg'><mrow><mi>&#967;</mi><mover accent='true'><mi>&#967;</mi><mo>&#175;</mo></mover><mo>&#8594;</mo><msup><mi>X</mi><mo>*</mo></msup><mo>&#8594;</mo><mi>g</mi><mi>g</mi></mrow></math> is <italic>p</italic>-wave suppressed and resonantly enhanced at <math altimg='si27.svg'><mrow><msub><mi>m</mi><mi>&#967;</mi></msub><mo>&#8771;</mo><msub><mi>M</mi><mi>X</mi></msub><mo linebreak='goodbreak'>/</mo><mn>2</mn><mo linebreak='goodbreak'>=</mo><mn>3.46</mn><mspace width='0.33em' /><mtext>GeV</mtext></mrow></math>. We show that the most important phenomenological uncertainty is no longer the existence of <italic>X</italic>(6900), which is now highly significant, but the interpretation of its line shape and hence the extraction of its mass, width, and gluonic coupling. We argue that a small set of numerical figures is essential for a publication-quality presentation: the direct-detection exclusion plane, the resonance line shape, and the resonant annihilation line shape.

Iran

Majid Monemzadeh

<italic>X</italic>(6900) as a hadronic mediator of a gluonic dark-matter portal

We investigate the gravitational deflection of charged massive particles by a charged galactic wormhole supported by the Sofue dark matter density profile [56], <math altimg='si2.svg'><mrow><mi>&#961;</mi><mrow><mo>(</mo><mi>r</mi><mo>)</mo></mrow><mo linebreak='goodbreak'>=</mo><msub><mi>&#961;</mi><mn>0</mn></msub><msup><mi>e</mi><mrow><mo>&#8722;</mo><mi>r</mi><mo>/</mo><msub><mi>r</mi><mn>0</mn></msub></mrow></msup><mo>,</mo></mrow></math> where <italic>&#961;</italic> <inf>0</inf> and <italic>r</italic> <inf>0</inf> denote the central dark matter density and the characteristic scale radius, respectively. This phenomenological profile models the dark matter distribution in galactic halos and acts as the matter source sustaining the wormhole geometry. Extending our previous investigation of light rays and neutral massive particles in this spacetime [57], we study the combined effects of gravitational and electromagnetic interactions on the motion of charged massive particles. The deflection angle is computed independently using the Rindler&#8211;Ishak method based on the Jacobi metric and the Gauss&#8211;Bonnet theorem, enabling a systematic comparison between the two approaches. We find that the predictions of the two methods become nearly indistinguishable in the relativistic regime, whereas the differences at lower velocities arise from the velocity-dependent correction terms proportional to <math altimg='si3.svg'><mrow><mo>(</mo><mn>1</mn><mo linebreak='goodbreak'>&#8722;</mo><msup><mi>v</mi><mn>2</mn></msup><mo>)</mo></mrow></math> in the trajectory equation. We further perform backward relativistic ray tracing of optically thin accretion flows around the charged galactic wormhole to investigate its shadow and photon-ring structure. Although the shadow remains circular because of the spacetime&#8217;s spherical symmetry, its radius, photon-ring structure, and intensity distribution exhibit a nontrivial dependence on the wormhole charge, providing potential observational signatures of charged galactic wormholes. These results provide new insights into the interplay between gravitational and electromagnetic interactions in charged wormhole spacetimes and their observational manifestations.

India

Md Khalid Hossain

Humaun Kabir Laskar

Farook Rahaman

Aritra Sanyal

Charged galactic wormholes: Shadow imaging, accretion disks, and charged-particle deflection

Two Higgs doublets models with a scalar potential that breaks electroweak symmetry spontaneously can have either one or two local minima. While potentials with one minimum can have a decoupling regime where all the new scalars are heavy, we show that, for potentials with <italic>two</italic> local minima, the masses of all the scalars are bounded if the dimensionless quartic couplings obey perturbativity constraints.

Spain

José M. Camacho

Italy

Carlos Miró

Miguel Nebot

Tomás Tobarra

On vacua and bounded masses in the general 2HDM

In this paper, we investigate the Krylov complexity of the massless scalar field in the massive Bañados-Teitelboim-Zanelli (BTZ) black hole with quantum fluctuation. This fluctuation is considered to be triggered by the Hawking radiation, which indeed influences the behavior of the time evolution of Krylov complexity. Our results show that Lanczos coefficients <italic>b<inf>n</inf></italic> exhibit a black hole mass <italic>M</italic>-dependent staggering, <math altimg='si7.svg'><mrow><mstyle mathvariant='normal'><mi>&#916;</mi></mstyle><mover><mrow><msub><mi>b</mi><mi>N</mi></msub></mrow><mo stretchy='true'>&#8254;</mo></mover><mo>&#8764;</mo><mi>M</mi></mrow></math>, inducing oscillatory complexity dynamics that intensify with lower mass black holes. Moreover, the graviton mass <italic>m</italic> is found to suppress the oscillations but accelerate the late-time exponential growth <math altimg='si8.svg'><mrow><mi>K</mi><mrow><mo>(</mo><mi>t</mi><mo>)</mo></mrow><mo>&#8764;</mo><msup><mi>e</mi><mrow><msub><mi>&#955;</mi><mi>K</mi></msub><mi>t</mi></mrow></msup></mrow></math>. This study indicates that the Krylov complexity could be a sensitive probe to distinguish massless and massive gravity via operator growth signatures.

China

Yu-Ting Zhou

Xiao-Mei Kuang

Krylov complexity of a massless scalar field in massive BTZ black hole with quantum fluctuation

We study a quantum-corrected version of the Oppenheimer–Snyder (OS) spacetime in which classical collapse is modified by a single effective parameter <italic>&#945;</italic>, representing small quantum gravitational effects in a phenomenologically rescaled form. Although the correction enters through a term proportional to <italic>&#945;M</italic> <sup>2</sup>/ <italic>r</italic> <sup>4</sup>, it produces noticeable changes in the strong-field region while leaving weak-field physics almost unchanged. One of the main results is the appearance of a minimum mass for horizon formation, <math altimg='si19.svg'><mrow><msub><mi>M</mi><mi>min</mi></msub><mo>&#8764;</mo><mrow><mo>(</mo><mn>0.5</mn><mo linebreak='badbreak'>&#8722;</mo><mn>1.2</mn><mo>)</mo></mrow><mspace width='0.16em' /><msub><mi>M</mi><mo>&#8857;</mo></msub></mrow></math> for the effective parameter <italic>&#945;</italic>&#8239;&#8776;&#8239;0.22. We emphasize that this <italic>M</italic> <inf>min</inf> is expressed in terms of the effective parameter <italic>&#945;</italic>; in terms of the bare loop quantum gravity (LQG) parameter, <math altimg='si20.svg'><mrow><msub><mi>M</mi><mi>min</mi></msub><mo>&#8764;</mo><msub><mi>M</mi><mtext>Pl</mtext></msub><mo>&#8764;</mo><msup><mn>10</mn><mrow><mo>&#8722;</mo><mn>5</mn></mrow></msup></mrow></math> g, consistent with the theoretical expectation. Orbital properties are also slightly modified, with the innermost stable circular orbit (ISCO) shifting to <math altimg='si21.svg'><mrow><msub><mi>r</mi><mrow><mrow><mi mathvariant='normal'>I</mi></mrow><mi>S</mi><mi>C</mi><mi>O</mi></mrow></msub><mo linebreak='goodbreak'>=</mo><mn>6</mn><mi>M</mi><mo linebreak='goodbreak'>&#8722;</mo><mi>&#945;</mi><mo linebreak='goodbreak'>/</mo><mrow><mo>(</mo><mn>12</mn><mi>M</mi><mo>)</mo></mrow></mrow></math>, giving corrections of order <math altimg='si22.svg'><mrow><mo>&#8764;</mo><msup><mn>10</mn><mrow><mo>&#8722;</mo><mn>4</mn></mrow></msup></mrow></math> for stellar black holes and <math altimg='si23.svg'><mrow><mo>&#8764;</mo><msup><mn>10</mn><mrow><mo>&#8722;</mo><mn>10</mn></mrow></msup></mrow></math> for Sgr A*. In the same region, radial epicyclic frequencies change by about <math altimg='si24.svg'><mrow><mn>5</mn><mo>%</mo><mo>&#8722;</mo><mn>10</mn><mo>%</mo></mrow></math> near <italic>r</italic>&#8239;&#8764;&#8239;5 <italic>M</italic>, while vertical modes are less affected, leading to small but structured shifts in quasi-periodic oscillation (QPO) behavior. Using four X-ray binaries (GRS 1915+105, XTE J1550-564, XTE J1859+226, GRO J1655-40), we find a consistent range <math altimg='si18.svg'><mrow><mi>&#945;</mi><mo linebreak='goodbreak'>=</mo><mn>0.22</mn><mo>&#177;</mo><mn>0.10</mn></mrow></math>, with masses between <math altimg='si25.svg'><mrow><mn>5.4</mn><mo linebreak='goodbreak'>&#8722;</mo><mn>12.4</mn><mspace width='0.16em' /><msub><mi>M</mi><mo>&#8857;</mo></msub></mrow></math> and emission radii <math altimg='si26.svg'><mrow><mi>r</mi><mo linebreak='goodbreak'>/</mo><mi>M</mi><mo>&#8764;</mo><mn>5.5</mn><mo linebreak='goodbreak'>&#8722;</mo><mn>8.6</mn></mrow></math>. The observed QPO frequencies, lying in the range <math altimg='si27.svg'><mrow><mn>100</mn><mo linebreak='goodbreak'>&#8722;</mo><mn>450</mn></mrow></math> Hz, are well reproduced within this framework. Weak-field tests such as S2 and Mercury still allow <math altimg='si28.svg'><mrow><msub><mi>f</mi><mrow><mrow><mi mathvariant='normal'>s</mi></mrow><mi>p</mi></mrow></msub><mo linebreak='goodbreak'>=</mo><mn>1.10</mn><mo>&#177;</mo><mn>0.19</mn></mrow></math>, leaving room for these small strong-gravity corrections. This work demonstrates that while bare Planck-scale corrections are unobservable, effective rescaled parameters that encode the collective effect of quantum gravity can be constrained by current astrophysical observations.

South Africa

Azerbaijan

Jordan

A. Errehymy

Morocco

Y. Khedif

M. Daoud

Uzbekistan

B. Turimov

S. Usanov

F. Turaev

Z. Yasakov

Joint constraints on quantum Oppenheimer–Snyder black holes from high-frequency quasi-periodic oscillations in X-ray binary systems and the S2 star orbit

Carrollian amplitudes are scattering amplitudes of massless particles written in position space at null infinity. We study various aspects of Carrollian amplitudes for gauge theory and gravity at loop level using primarily the modified Mellin prescription. Finite one-loop four-point Carrollian amplitudes in gauge theory are shown to maintain an analytic structure similar to tree level results. We compute the one-loop four-point Carrollian MHV amplitudes in planar 𝒩 = 4 super Yang-Mills theory, which are expressed as differential operators acting on tree level Carrollian amplitudes. This result is generalized to all loop orders using the Bern-Dixon-Smirnov (BDS) formula. Similar structures are observed at one-loop for Carrollian MHV amplitudes in 𝒩 = 8 supergravity. We next consider 2 → 2 scattering of massless scalars via gravitational interactions in the eikonal regime and show that the corresponding Carrollian amplitudes exhibit logarithmic behavior in the ‘Carroll time’ $\textit{u}$. We compute the discontinuities of these Carrollian amplitudes up to 𝒪($\textit{G}$ ) and show that they are descendants of Carrollian Born amplitudes. We observe similar logarithmic behavior in Carrollian amplitudes associated with the one-loop scalar box diagram. The dependence of this amplitude on dual scaling dimensions also differs from standard tree level results. Finally, we further study the infrared (IR) divergences of Carrollian amplitudes in massless scalar QED, gravity, and Yang-Mills theory. We show that Carrollian amplitudes in these theories naturally factorize, allowing us to provide an IR-safe definition for these objects.

United Kingdom

Vijay Nenmeli

Bin Zhu

On Carrollian loop amplitudes for gauge theory and gravity

The all-sky very-high-energy (10 − 10 GeV) atmospheric muon flux measured by IceCube shows a spectral hardening at the highest energies, indicating the presence of a prompt component. IceCube has also measured the atmospheric muon neutrino flux at high energy. However, since this flux is dominated by astrophysical neutrinos, only an upper bound can be placed on the prompt atmospheric $${\nu}_{\mu }+{\overline{\nu }}_{\mu }$$ contribution. In this work, we provide a new evaluation of the prompt atmospheric muon flux including an intrinsic charm component in the cosmic ray-air interactions. The latter enhances the forward production of $${\overline{D} }^{0}$$ , $\textit{D}$ , and Λ , which subsequently decay into final states containing muons and muon neutrinos. We show how the increase in the prompt muon flux due to intrinsic charm is accompanied by a corresponding enhancement in the prompt muon neutrino flux. We implement different intrinsic charm production models in MCEq to calculate the resulting lepton fluxes. We discuss the challenges of achieving predictions that are simultaneously consistent with both IceCube’s high-energy atmospheric muon flux measurements and IceCube upper bounds on the prompt muon neutrino flux, and we quantify the resulting discrepancies. As possible solutions, we explore scaling of the unflavored meson contributions to the prompt atmospheric muon flux to assess how such adjustments can reconcile these differences. The tensions emphasized in our work call for a refinement of the hadronic interaction models, especially the production of unflavored mesons, and for new experimental data sensitive to unflavored meson and heavy flavor production with reliable estimates of the associated uncertainties. We suggest that the energy and zenith angle dependence of muon and neutrino flux ratios from future neutrino telescope measurements may help to disentangle different scenarios.

United States

Laksha Das

Diksha Garg

Maria Garzelli

Mary Reno

Günter Sigl

The role of charm and unflavored mesons in prompt atmospheric lepton fluxes

We present the first calculation of the next-to-leading-order electroweak corrections to vector-boson scattering into doubly polarised Z bosons at the LHC in the fully leptonic decay channel. The production and decay of the two polarised Z bosons are consistently modelled in the double-pole approximation, separating polarisation states at the amplitude level and including factorisable real and virtual electroweak corrections. Doubly polarised and unpolarised signals are investigated and confronted with off-shell results. A broad analysis, including results at integrated and differential level, is carried out in a realistic, CMS-inspired fiducial setup. Our study paves the way to upcoming analyses with LHC Run-3 and High-Luminosity data as well as to further phenomenological investigations.

Ansgar Denner

Robert Franken

Santiago Chavez

Daniele Lombardi

Giovanni Pelliccioli

The cleanest of them all: NLO electroweak corrections to vector-boson scattering into doubly polarised ZZ pairs at the LHC